Knowledge catalogue
arXiv:2603.27909v2 Announce Type: replace-cross Abstract: Car-following behavior is fundamental to traffic flow theory, yet traditional models often fail to capture the stochasticity of naturalistic d
I know I know, it's great, we know. I've been working on tweaking inference engines for a week now and it's been one shotting most of my vague prompts without any issues. It will even write tests and
arXiv:2609.12162v1 Announce Type: cross Abstract: Draft-verify-revise is a common LLM orchestration pattern for scaling inference-time compute. One LLM drafts, a second critiques the draft and provide
arXiv:2609.11998v1 Announce Type: new Abstract: Diffusion language models decode tokens in parallel, but their bidirectional denoiser rules out the naive key--value (KV) cache behind fast autoregressi
We are excited to share that Google Cloud was named a Leader and received the highest score in the ‘current offering’ category in the Forrester Wave™: Public Cloud Platforms, Q3 2026 report, which exa
arXiv:2609.12408v1 Announce Type: new Abstract: Hazardous-freight operations at petrochemical logistics nodes are safety-critical for intelligent transportation systems, yet existing vision benchmarks
arXiv:2403.05864v3 Announce Type: replace Abstract: In human-centric Cyber-Physical Systems (CPS), personalized Deep Reinforcement Learning (DRL) agents must share fine-grained control actions with cl
I finished my home inference server. First I tried Lenovo p620 workstation and while it’s a good value overall it pissed me off with a ton of proprietary Lenovo shit to deal with and I return it in th
It’s always bothered me that after fine-tuning a model for a project, there isn’t a particularly easy way to host it without either running it locally and keeping a GPU on 24/7 or paying for an entire
arXiv:2609.11713v1 Announce Type: cross Abstract: Artificial neural networks rely on vector-matrix multiplications (VMMs), whose implementation in von Neumann architectures is dominated by costly data
arXiv:2609.11916v1 Announce Type: new Abstract: Camera traps often run in the field on edge hardware with limited or no connectivity, making small, locally-deployable vision-language models (VLMs) --
Hey everyone — I’m building TensorSharp, an open-source LLM inference engine. Here are the latest DeepSeek V4.1 Flash GGUF results using its native ggml_cuda backend. Setup: 8× NVIDIA A40, layer split
arXiv:2609.11314v1 Announce Type: new Abstract: Traffic flow prediction is a core supporting technology for intelligent transportation systems. It uses historical data to infer future traffic dynamics
arXiv:2609.11281v1 Announce Type: cross Abstract: Learning and decision-making in animals are often modeled as Bayesian processes, where sensory evidence is integrated with prior beliefs to guide beha
arXiv:2609.09229v1 Announce Type: cross Abstract: Additional inference compute can increase the number of correctly resolved security-assurance tasks, but repeated success, unique coverage, accepted e
arXiv:2609.10830v1 Announce Type: new Abstract: When a language model finds a sentence unusually cheap to predict, it is tempting to conclude that the sentence was in its training data. Almost every p
arXiv:2609.10658v1 Announce Type: new Abstract: Activation steering provides a lightweight way to control large language models (LLMs) by modifying their hidden activations at inference time. Among th
Hey gang, im wondering if you in theory could use ngrams as seen with Qwen 3.8 Flash or DS4.1 in order to dynamically train the model? Normally the ngram embeddings behave similar to a lookup table of
arXiv:2609.11133v1 Announce Type: new Abstract: Datacenter GPU power is the binding constraint on LLM serving capacity, and production serving has shifted to prefill/decode (PD) disaggregation. Deploy
arXiv:2609.11255v1 Announce Type: cross Abstract: Radiomap blind prediction infers radiomaps from observable representations of the propagation environment and base station (BS) configuration without
arXiv:2609.11548v1 Announce Type: new Abstract: Autoregressive video world models enable interactive, long-horizon exploration, but flexible control remains challenging. Exploring a source video from
AI inference chipmaker d-Matrix today announced it will use NVIDIA NVLink Fusion to connect its next-generation Raptor XPUs to NVIDIA’s AI infrastructure platform — joining a growing roster of ecosyst
arXiv:2609.10018v1 Announce Type: new Abstract: EdgeAI systems are increasingly employing computer vision applications to enable intelligent, on-device decision-making in real-time. However, these dep
arXiv:2609.09634v1 Announce Type: new Abstract: Large networks and ensembles often lead medical image segmentation challenges, but their storage and inference demands complicate deployment. We present
arXiv:2609.09691v1 Announce Type: new Abstract: When training data are limited, increasing parameter count is not the only way to improve language-model performance. A small parameter set, when repeat
arXiv:2609.10054v1 Announce Type: cross Abstract: Orukeet replaces half of an adapted Parakeet encoder's temporal filters with 12,288 fitted Gabor kernels, freezes these replacements, and trains the r
arXiv:2609.07901v1 Announce Type: new Abstract: Weight quantization largely determines the economics of serving open-weight LLMs. Its costs are usually assessed with capability benchmarks, on which 4-
arXiv:2609.09764v1 Announce Type: new Abstract: Social intelligence enables agents to read social context, infer intent, and adapt over sustained dialogue. As language models become autonomous collabo
arXiv:2609.09989v1 Announce Type: new Abstract: A natural way to cut reasoning-model inference cost is to repeatedly probe a single partial trajectory for its current answer and stop once probes agree
arXiv:2609.10297v1 Announce Type: new Abstract: GUI agents accumulate high-resolution screenshots as the trajectory unfolds, increasing inference latency and memory usage. Training-free visual token p
arXiv:2609.09799v1 Announce Type: cross Abstract: Crystal structure prediction (CSP) is fundamental to computational materials discovery. Generative models including CDVAE, DiffCSP, FlowMM, and Crysta
arXiv:2609.07785v1 Announce Type: new Abstract: An LLM-agent leaderboard invites a familiar inference: an agent ranked above another is the better agent. Public evaluation logs may not support that co
arXiv:2609.08162v1 Announce Type: new Abstract: How can vision-language-action (VLA) models adapt to new environments where world dynamics shift? While recent research has combined world modeling and
mentria.ai is a browser inference engine I've been building solo, from scratch in WebGPU/WGSL. This week it crossed a milestone I had been chasing for a while: a 27B one-bit model answering at up to 3
arXiv:2609.05484v1 Announce Type: new Abstract: Deep learning registration methods routinely stack two kinds of enhancement on a base network: architectural additions such as affine pre-alignment stag
arXiv:2609.07094v1 Announce Type: new Abstract: Reasoning is a promising route to the generalization that autonomous driving requires in the long tail, as it can infer how the elements of a scene depe
arXiv:2609.06133v1 Announce Type: new Abstract: Sequential inference on small devices requires a model to retain useful history without repeatedly processing a long input record. A Recurrent Tsetlin M
arXiv:2609.06880v1 Announce Type: new Abstract: Reasoning over language instructions in embodied tasks such as robotics often requires understanding spatial relations from a speaker's situated perspec
arXiv:2512.21653v2 Announce Type: replace-cross Abstract: Speech codecs are usually optimized for waveform fidelity, allocating bits to acoustic detail that can be inferred from linguistic structure.
arXiv:2609.07037v1 Announce Type: new Abstract: Activation steering has emerged as a lightweight, inference-time approach to control the behavior of Large Language Models (LLMs). However, traditional
arXiv:2603.18493v2 Announce Type: replace Abstract: Streaming 3D reconstruction maintains a persistent latent state that is updated online from incoming frames, enabling constant-memory inference. A k
arXiv:2609.06004v1 Announce Type: new Abstract: Quantitative spatial reasoning in visual-language models (VLMs) aims to infer spatial distances and directional relationships among objects in 3D space
arXiv:2609.08306v1 Announce Type: cross Abstract: We introduce HoneyRoute, an inference-serving layer that detects whether an incoming request is malicious and, if so, routes it to a dedicated honeypo
arXiv:2603.01875v4 Announce Type: replace-cross Abstract: Knowledge distillation (KD) is widely used to compress and post-train large language models (LLMs), yet many existing frameworks execute teach
arXiv:2602.13174v2 Announce Type: replace Abstract: Partial differential equation (PDE) models frequently contain unknown functional terms that cannot be measured directly, limiting their predictive u
arXiv:2609.09009v1 Announce Type: new Abstract: Denoising Diffusion Probabilistic Models (DDPMs) generate samples by starting from noise and repeatedly denoising while keeping each update close to the
arXiv:2609.07627v1 Announce Type: cross Abstract: AI agents sometimes act aligned when they infer they are being tested, and differently when not. We argue this is not an anomaly but what current trai
arXiv:2609.05436v1 Announce Type: cross Abstract: Protein scaffold gap filling is an important computational task in protein sequence reconstruction, where missing amino acid regions must be inferred
arXiv:2609.06288v1 Announce Type: new Abstract: Training-free image editing steers diffusion or flow-matching generative models at inference time by modifying prompt-conditioned denoising velocities.
arXiv:2609.05676v1 Announce Type: new Abstract: Language models adapted on private text are often served through APIs, so privacy leakage occurs through generated outputs rather than exposed weights.
arXiv:2609.09145v1 Announce Type: new Abstract: We introduce Point4D, a feed-forward model for 4D reconstruction of long-range video sequences. Point4D is able to reliably infer dense per-point 3D tra
arXiv:2609.05896v1 Announce Type: cross Abstract: Link sign prediction (LSP) aims to infer the positive or negative polarity of unobserved links in signed networks. Signed Graph Neural Networks (SGNNs
arXiv:2609.09148v1 Announce Type: new Abstract: Generalist robot policies carry broad manipulation priors from large-scale data, but specializing them to a new task remains the deployment bottleneck.
arXiv:2609.06061v1 Announce Type: new Abstract: Termination-based constrained reinforcement learning is attractive for safety-critical robotic deployments: it avoids online optimization at inference,
arXiv:2609.06289v1 Announce Type: cross Abstract: As large language models (LLMs) are increasingly deployed in alignment-sensitive contexts, activation steering has emerged as a lightweight, inference
arXiv:2609.06473v1 Announce Type: new Abstract: Inference-time activation steering enables behavioral control of large language models without parameter modification, while post-training quantization
arXiv:2603.02935v2 Announce Type: replace Abstract: Offline meta-reinforcement learning seeks to learn a policy that generalizes to new related tasks online. Context-based methods infer a task represe
arXiv:2609.08367v1 Announce Type: new Abstract: Test-time adaptation (TTA) has emerged as a prominent strategy for adapting vision-language models to distribution shifts during inference. We conduct a
arXiv:2609.08703v1 Announce Type: cross Abstract: Text-to-speech systems often face a trade-off between natural prosody and efficient inference: higher perceptual quality typically comes at increased
arXiv:2609.08659v1 Announce Type: new Abstract: Onboard object detection in Earth observation is constrained by limited computational resources and the absence of fully corrected imagery. While convol